RESEARCH ARTICLE | PHARMACOLOGY

Insulin-­inspired hippocampal neuron–targeting technology

for protein drug delivery
Noriyasu Kameia,1 , Kento Ikedaa, Yuka Ohmotoa, Seita Fujisakia, Ryusei Shirataa, Maya Makia, Mika Miyataa, Yuki Miyauchia, Nanaka Nishiyamaa,
Mana Yamadaa, Yuna Ohigashia, and Mariko Takeda-­Morishitaa

Affiliations are included on p. 11.

Edited by Michael Goldberg, Columbia University, New York, NY; received April 20, 2024; accepted September 3, 2024

Hippocampal neurons can be the first to be impaired with neurodegenerative disorders,

including Alzheimer’s disease (AD). Most drug candidates for causal therapy of AD Significance
cannot either enter the brain or accumulate around hippocampal neurons. Here, we
genetically engineered insulin-­fusion proteins, called hippocampal neuron–targeting There are many hurdles in
(Ht) proteins, for targeting protein drugs to hippocampal neurons because insulin tends developing biological drugs
to accumulate in the neuronal cell layers of the hippocampus. In vitro examinations against neurodegenerative
clarified that insulin and Ht proteins were internalized into the cultured hippocam- disorders, such as Alzheimer’s
pal neurons through insulin receptor–mediated macropinocytosis. Cysteines were key disease (AD). One is the blood–
determinants of the delivery of Ht proteins to hippocampal neurons, and insulin B brain barrier (BBB), which strictly
chain mutant was most potent in delivering cargo proteins. In vivo accumulation of Ht limits the entering of exogenous
proteins to hippocampal neuronal layers occurred after intracerebroventricular admin-
molecules, including drugs. This
istration. Thus, hippocampal neuron–targeting technology can provide great help for
has been overcome by various
developing protein drugs against neurodegenerative disorders.
strategies with huge efforts and
hippocampal neuron | insulin | drug delivery | fusion protein | Alzheimer’s disease advances during the last few
decades. Another significant
Neurodegenerative disorders increase with age. In particular, dementia, such as Alzheimer’s barrier is the inefficient
disease (AD), not only affects the patients but also their family and caregivers. To extend accumulation of drugs to the
healthy life expectancy, effective preventive and therapeutic strategies against neurode­ desired target site in the brain,
Downloaded from https://www.pnas.org by 190.6.179.10 on January 10, 2025 from IP address 190.6.179.10.

generative disorders must be developed. Clinically, a limited number of medications are in particular, the hippocampus,
available for AD, including donepezil, memantine, and galantamine, which symptomat­ which is damaged in
ically prevent decrease in neuronal activity and communication (1). Recently, an antibody neurodegenerative disorders,
drug (lecanemab) for removing pathogenic amyloid β (Aβ) was approved in the United including AD. Here, we genetically
States and Japan (2, 3). However, the use of antibody drugs is limited for mild cognitive engineered insulin-fusion proteins
impairment and early stages of AD because there is no evidence that they are effective on
[called hippocampal neuron–
the progressive stage of dementia, with severe pathological conditions and memory loss.
Furthermore, their effectiveness is controversial between specialists and researchers. targeting (Ht) proteins] and
Most candidates are currently proposed as drugs for treating AD (Fig. 1A). Although provided evidence that Ht proteins
antibodies are thought to remove soluble Aβ oligomers and protofibrils from the brain could be targeted to hippocampal
and blood, inhibitors for Aβ-producing enzymes from amyloid precursor protein, such neurons in vitro and in vivo. Our
as β-secretase (BACE1), can prevent the generation of Aβ. Nucleic acids, including anti­ method will help in the
sense oligos and small interfering RNA (siRNA), can reduce Aβ production by silencing development of protein drugs for
BACE1 gene expression. Recently, nucleic acids were tested for silencing the gene encoding neurodegenerative disorders.
microtubule-associated protein tau, another pathogen in AD (4, 5). Neurotrophins, such
as brain-derived neurotrophic factor, may be used to recover loss of neuronal function Author contributions: N.K., K.I., Y. Ohmoto, S.F., R.S.,
and activity under pathological conditions (6–9). It is important that the drugs should M. Maki, M. Miyata, Y.M., N.N., M.Y., and Y. Ohigashi
work in or around neuronal cells at the “hippocampus” because hippocampal neurons are designed research; N.K., K.I., Y. Ohmoto, S.F., R.S.,
M. Maki, M. Miyata, Y.M., N.N., M.Y., and Y. Ohigashi
initial points for the production and secretion of Aβ, as well as the phosphorylation or performed research; N.K., K.I., Y. Ohmoto, S.F., R.S.,
propagation of tau. M. Maki, M. Miyata, Y.M., N.N., M.Y., and Y. Ohigashi
analyzed data; M.T.-­M. gave suggestions; and N.K., K.I., Y.
In recent decades, pharmaceutical companies have been struggling to develop treat­ Ohmoto, S.F., R.S., M. Maki, M. Miyata, Y.M., N.N., M.Y., Y.
ments, including antibody drugs, for the causal treatment of AD. Most potential Ohigashi, and M.T.-­M. wrote the paper.

candidates have been dropped out from clinical trials, and development was switched The authors declare no competing interest.

for next drug candidates, without considering their pharmacokinetic problems. Because This article is a PNAS Direct Submission.

only few molecules (transporter substrates like glucose and hydrophobic small mole­ Copyright © 2024 the Author(s). Published by PNAS.
This article is distributed under Creative Commons
cules) can permeate through the blood–brainbarrier (BBB), most drugs cannot pene­ Attribution-­NonCommercial-­NoDerivatives License 4.0
trate the brain parenchyma (Fig. 1B). Pharmaceutical research has little addressed the (CC BY-­NC-­ND).
transport of antibody drugs from systemic circulation to the brain. Various promising 1
To whom correspondence may be addressed. Email:
[email protected].
techniques for overcoming and bypassing the BBB have been established (10–12). The
This article contains supporting information online at
most well-known strategy is based on antibodies against transferrin receptor (TfR), https://www.pnas.org/lookup/suppl/doi:10.1073/pnas.​
which is abundantly expressed on the surface of brain microvascular endothelial cells. 2407936121/-­/DCSupplemental.
Cargo drugs fused to the TfR antibody can be delivered to the brain (13–18). Another Published September 30, 2024.

PNAS 2024 Vol. 121 No. 41 e2407936121 https://doi.org/10.1073/pnas.2407936121 1 of 12

 A

B
Downloaded from https://www.pnas.org by 190.6.179.10 on January 10, 2025 from IP address 190.6.179.10.

Fig. 1.   Schematic conceptual illustration of the study. (A) Need for targeted delivery of drugs, including antibodies, neurotrophic factors, and nucleic acids, to
hippocampal neurons for protection and repair in neurodegenerative disorders, such as dementia (AD). (B) Obstacles for transport of drugs to the brain. The
BBB prevents entry of exogenous molecules, even therapeutic drugs. Drugs are required to reach deep target regions of the brain, such as the hippocampus,
through the brain parenchyma. (C) Unique strategy to deliver protein-­based drugs to the hippocampus via fusion with insulin, which potentially accumulates on
hippocampal neurons. Our goal is to effectively deliver insulin-­fusion protein therapeutics from administration sites (nose or systemic circulation) to hippocampal
neurons. Illustrations were created with BioRender.com.

useful option is nose-to-brain delivery, in which drugs intrana­ drugs for dementia; however, such approaches have not been
sally administered can directly move to the brain without sys­ studied yet.
temic absorption and subsequent transport through BBB Insulin is a bioactive peptide that performs many functions in
(19–22). These techniques will accelerate drug development for the body via receptor stimulation, which triggers phosphorylation
AD and other neurodegenerative disorders. However, as of downstream signaling molecules, such as insulin receptor sub­
described above, one of the major therapeutic targets for these strate (IRS), phosphoinositide 3-kinase (PI3K), and protein kinase
diseases related to memory loss and cognitive dysfunction is the B (Akt) (23). Over several decades, the IR on the BBB was referred
hippocampus. Therefore, the technology for specifically accu­ to as a potential target for drug delivery to the brain, but transcy­
mulating them to hippocampal neuronal cells is necessary with tosis via the IR on the brain endothelial cells might not be medi­
or after brain delivery to maximize the efficacy of the potential ated by normal IR signaling (24). On the other hand, recent

2 of 12 https://doi.org/10.1073/pnas.2407936121 pnas.org
 studies have clarified that insulin can enhance memory and learn­ quantitative data with histograms and geometric mean fluores­
ing after intranasal administration (25, 26). We recently found cence intensity, based on flow cytometry, respectively. Among
that insulin could be delivered efficiently to the brain via intranasal brain cells, neuronal HT22 cells are highly capable of FITC-insulin
coadministration with cell-penetrating peptide (penetratin), as uptake compared with glial C6 cells. In contrast to the known
the epithelial permeation enhancer, and eventually distributed property of insulin passing through BBB, FITC-insulin was less
specifically around the neuronal cell layers in the mouse hip­ taken up by bEnd.3 cells. Because IR is widely expressed in the
pocampus (27). We hypothesized that the disposition of insulin body, FITC-insulin was effectively taken up by 3T3 cells. As
in the brain makes it eligible for efficient drug delivery to the shown in Fig. 2G, FITC-insulin was more effectively taken up by
hippocampus and neuronal cells (Fig. 1C). Therefore, we devel­ HT22 cells than neuroblastoma Neuro2a (N2a) cells, suggesting
oped targeting technology to deliver protein-based drugs, such as specific accumulation of insulin to the hippocampal neurons.
antibodies and neurotrophic factors, to hippocampal neurons, Examination of temperature sensitivity (Fig. 2 H and I) demon­
based on genetic engineering of fusion proteins. First, we evaluated strated that internalization of FITC-insulin was completely
the characteristics of insulin in the brain to estimate its potential abolished at low temperature (4 °C), suggesting that insulin inter­
as a drug carrier. Second, we generated plasmid vectors for bacte­ nalization involves energy-dependent pathways, including endo­
rial expression of proteins fused to insulin (insulin A chain, B cytosis. Punctate cellular fluorescence (Fig. 2D) indicated
chain, and full-length) and tested the ability of insulin-fusion internalization of FITC-insulin via endocytosis.
proteins to accumulate in hippocampal neurons. The fusion pro­ We further examined IR expression on the cells to clarify the
teins were named “hippocampal neuron–targeting (Ht) proteins.” relationship with cellular uptake of insulin. IR was most highly
Third, we studied the mechanisms associated with specific local­ expressed in the 3T3 cells, and the order of expression level in the
ization of Ht proteins in the brain under various experimental brain-related cells was HT22, C6, and bEnd.3 (Fig. 2 J and K).
conditions and with insulin mutants. Furthermore, we demon­ According to the company’s datasheet of primary antibody, the
strated the in vivo performance of Ht proteins in effective and main band for IR should appear around 95 kDa, but in mouse
specific delivery to hippocampal neurons after intracerebroven­ tissues, the band at 72 kDa seemed correct. This trend was similar
tricular (ICV) administration in mice and tested the brain delivery with that of cellular internalization capacity (Fig. 2F), implying
potential through noninvasive intranasal administration. This that insulin accumulated in hippocampal neurons via receptor
study provides evidence that a specific targeting approach with binding. Because receptor binding of some growth factors may
insulin-inspired fusion proteins can enhance accumulation of have cross-effect with insulin, we examined insulin-like growth
protein drugs and other molecules in hippocampal neurons. Our factor 1 receptor (IGF-1R) (29, 30). However, IGF-1R was not
findings have great application in brain neurodegenerative disorder detected in any cell lysate used in this study (SI Appendix,
therapy. Fig. S1C).
Thus, insulin can deliver other molecules into hippocampal neu­
Results rons. We tested the ability of insulin to deliver the low molecular
Downloaded from https://www.pnas.org by 190.6.179.10 on January 10, 2025 from IP address 190.6.179.10.

fluorescent dye, FITC, via ICV administration of FITC-insulin.

In Vivo and In Vitro Disposition of Insulin around the Hippo­ Fluorescence was observed in the hippocampal neuronal layer after
campus after Efficient Brain Delivery. In our previous study, ICV administration of FITC-insulin, whereas FITC-dextran (4.4
insulin seemed to be localized on hippocampal neuronal layers kDa, FD-4) tagged with FITC could not be localized in neuronal
after enhanced nose-­to-­brain delivery (27). To discuss the potential cells (SI Appendix, Fig. S2). The results suggest that insulin can deli­
of insulin in delivering other drugs to hippocampal neurons, ver other molecules, possibly protein drugs, through conjugation.
we first evaluated the in vivo distribution property of insulin
after intranasal administration, based on immunohistological Genetic Engineering, Expression, and Characterization of Ht
examinations (Fig. 2A). The results reproduced our results that Proteins. We engineered fusion proteins of insulin and target proteins
insulin (green) could accumulate around hippocampal neuronal for delivery of protein drugs to hippocampal neurons. Luminescent
layers (Fig. 2B). Multiple staining results showed that insulin did and fluorescent proteins—NanoLuc (Nluc) and mNeonGreen
not colocalize with marker proteins for astrocytes and neuronal (mNG), respectively—were used to detect bio­distribution, brain
stem cells glial fibrillary acidic protein (GFAP) and nestin, penetration, and neuronal accumulation in the hippocampus
respectively). By contrast, colocalization with the neuronal cell (31, 32). pET vectors with genes encoding insulin (A chain, B
body marker microtubule-­associated protein 2 (MAP2) clarified chain, and full-­length proinsulin, including A and B chains and
that insulin was internalized by neuronal cells, suggesting that c-­peptide) and probe proteins (Nluc and mNG) were designed and
insulin was not only bound to the cell surface but was also taken transformed into bacteria for protein expression (Fig. 3A). Insulin
up by neuronal cells. Microglial marker Iba1 was not included in A– and B chain–fused proteins were named as HtA-­ and HtB
this study because insulin disposition was clearly different from proteins, respectively. Plasmids encoding full-­length proinsulin-­
typical microglial localization. fused proteins were tested being transformed into different
We further investigated the in vitro property of insulin by test­ bacteria; the fusion proteins expressed in BL21(DE3) and SHuffle
ing its cellular uptake by various cell types (Fig. 2C). Insulin accu­ were named HtC and HtD proteins, respectively. Here, SHuffle
mulation was compared mainly between C6 (glial cell model) and Escherichia coli was used to enhance the correct folding of disulfide-­
HT22 (neuronal cell model) cells. The brain microvascular model bonded proteins (33). All proteins tagged with a histidine–
(bEnd.3 cells) was used as insulin might have the potential to asparagine repeat (6× HN) were purified with nickel-­based gravity
transport through BBB (28). Fibroblasts NIH3T3 cells were used columns. The proteins were confirmed eluted in fraction No. 2,
as brain-unrelated control cells. Peripheral tissue-derived cells, and expression yields were partially reduced due to fusion with
such as HEK293T cells, with more abundant expression of IR insulin (Fig. 3 B and C). Luminescence activity and fluorescence
than neuronal cells (SI Appendix, Fig. S1 A and B), were excluded intensity were considerably decreased with addition of insulin to
from comparison of cellular uptake of Ht-fusion proteins. Fig. 2D the N terminus of the proteins (Fig. 3 D and E). We used western
shows confocal micrographs of cells treated with fluorescein iso­ blotting to confirm that insulin was fused to Nluc (Fig. 3F) and
thiocyanate (FITC)-insulin for 2 h. Fig. 2 E and F shows more mNG (Fig. 3G). Antibodies specific to insulin A or B chain could

PNAS 2024 Vol. 121 No. 41 e2407936121 https://doi.org/10.1073/pnas.2407936121 3 of 12

 A B

F
Downloaded from https://www.pnas.org by 190.6.179.10 on January 10, 2025 from IP address 190.6.179.10.

H I J K

Fig. 2.   In vitro and in vivo characterization of insulin as a carrier to hippocampal neurons. (A) Brief illustration of the in vivo study for evaluating the specificity
of insulin accumulation in the brain. For microscopic observation, brain specimens were isolated from mice receiving intranasal insulin (10 IU/kg) with mucosal
permeation enhancer (L-­penetratin, 2 mM). Frozen brain sections were treated with primary antibodies to insulin and various cell markers (MAP2, neuronal cell
body; GFAP, astrocytes; nestin, neuronal stem cells) and stained with secondary antibodies conjugated with fluorescent dyes. (B) Representative images (from
three mice) of the hippocampus after intranasal delivery of insulin. (C) Four types of cells were used for the in vitro study (NIH3T3 cells as unrelated control, bEnd.3
cells as BBB model, C6 cells as glial model, and HT22 cells as hippocampal neuron model). (D) Confocal micrographs of cells incubated with FITC-­insulin (10 μg/mL)
for 120 min at 37 °C. Prior to observation, the cells were washed thrice with glycine–HCl buffer and once with phosphate-­buffered saline (PBS) for optimal detection
of intracellular fluorescence. The bars indicate 20 μm. (E) Histograms generated from flow cytometry after incubation with FITC-­insulin (10 μg/mL) for 120 min
at 37 °C. (F) Geometric mean fluorescence intensity (MFI) calculated from the results of flow cytometry as the quantitative parameter. (G) Comparison of cellular
uptake of FITC-­insulin by HT22 and N2a cells. (H and I) Flow cytometry for examining energy dependence of cellular uptake of FITC-­insulin by comparing them
at biological (37 °C) and energy-­abolishing (4 °C) conditions. (J) Western blot analysis to compare the expression of IR between cells. Samples were loaded at 6
μg of total protein. β-­Actin was used as housekeeping control. (K) Relative expression levels of IR to β-­actin between cells based on western blotting.

4 of 12 https://doi.org/10.1073/pnas.2407936121 pnas.org
 A

F G
B D

C E

Fig. 3.   Production and characterization of Ht proteins. (A) Design of plasmid vectors for generating fusion proteins with insulin. The constructs were designed
for fusing target proteins with A chain (21 amino acids), B chain (30 amino acids), or full-­length insulin (B chain/c-­peptide/A chain, proinsulin form). Nluc luciferase
Downloaded from https://www.pnas.org by 190.6.179.10 on January 10, 2025 from IP address 190.6.179.10.

and mNG were chosen as experimental model proteins for detection in several assays. (B and C) The Ht proteins expressed in bacteria [BL21(DE3) or SHuffle T7
E. coli] were purified with nickel affinity columns through the HN-­repeat tag at the C terminus of fusion proteins. Ht proteins were mostly eluted in fraction #2.
(D and E) Luminescence or fluorescence intensity of Ht proteins (Nluc/mNG) was measured with a microplate reader at 100 pg/mL or 20 μg/mL, respectively.
(F and G) Western blotting of Ht-­Nluc and mNG, respectively. Samples were loaded at 75 ng/mL of purified protein. Poly vinylidene difluoride (PVDF) membranes
were treated with antibodies for Nluc, mNG, insulin A chain, and B chain. Arrows indicate the molecular weight of original Nluc or mNG.

react to HtA or HtB protein, respectively, and HtC-­ and HtD various cell types (3T3, C6, and HT22 cells) (Fig. 4 D and E);
proteins were detected with both anti-­insulin antibodies. Because importantly, mNG could be predominantly taken up by HT22
multiple bands were observed under nonreducing condition, fusion cells, especially when fused to insulin B chain (HtB-­mNG). These
proteins might be complexed intermolecularly. Analysis with high-­ in vitro estimations suggest that insulin, even fragments such as
performance liquid chromatography confirmed purified Ht-­fusion the B chain, can effectively deliver the proteins to hippocampal
proteins without any contamination (SI Appendix, Fig. S3). neurons.
The higher specificity of HtB- and HtD-mNG to HT22 cells
Comparison of Cellular Accumulation and Internalization was confirmed in HT22 and C6 cells cocultured Transwell assay.
Properties of Ht Proteins. To examine the hippocampal neuronal To avoid direction vias of Transwell (between upper and lower
specificity of Ht proteins, we compared qualitatively and quan­ chambers), one experiment was conducted with HT22 and C6
titatively their accumulation and internalization into different cell cultured in upper and lower chambers, respectively, and another
types with confocal microscopy and flow cytometry (Fig. 4A). As with the reverse order of cultures (SI Appendix, Fig. S4A). HtB-­
described for examination with FITC-­insulin, NIH3T3, bEnd.3, and HtD-mNG were more preferentially taken up by HT22 cells
C6, and HT22 cells were used in this examination (Fig. 2C). First, than C6 cells in both conditions (SI Appendix, Fig. S4B) The
we compared the internalization of all Ht-­fusion mNG in bEnd.3 uptake pattern of Ht-mNG was similar to that of FITC-insulin
(BBB model) and HT22 cells (hippocampal neuron model) and (SI Appendix, Fig. S4C). To elucidate the involvement of the recep­
found that insulin fusion effectively enhanced the cellular uptake tor binding of Ht-mNG in their cell accumulation specificity,
of mNG, particularly by HT22 cells (Fig. 4B). Intriguingly, intermolecular interactions between Ht-mNG and IR were ana­
even incomplete insulin (only A and B chains) could intensively lyzed with surface plasmon resonance (SPR) (Fig. 4F). In this
increase cellular internalization of mNG. The specificity of most experiment, purified IR, consisting of α and β subunits, which
effective HtB-­and complete HtD-­mNG was further tested with was cloned from HEK293T cells and expressed in BL21(DE3) E.
various cell types (Fig. 4C). Consistent with the results using coli, was immobilized on the CM5 sensorchip [706.4 response
FITC-­insulin (Fig. 2D), punctate fluorescence was observed in unit (RU)], and then, Ht-mNG were injected as analyte. A neg­
3T3, C6, and HT22 cells after treatment with HtB-­ and HtD-­ ligible change was observed in the sensorgram after injection of
mNG, but these proteins were slightly internalized into bEnd.3 mNG, confirming that mNG without insulin could not specifi­
cells. Flow cytometry–based quantification revealed that insulin cally bind to IR (Fig. 4G). By contrast, a concentration-dependent
fusion could potentially enhance the internalization of mNG into increase in the sensorgram was observed after all Ht-mNG was

PNAS 2024 Vol. 121 No. 41 e2407936121 https://doi.org/10.1073/pnas.2407936121 5 of 12

 A B

C D

E F

G
Downloaded from https://www.pnas.org by 190.6.179.10 on January 10, 2025 from IP address 190.6.179.10.

H J

Fig. 4.   In vitro cellular uptake and in vivo brain distribution of Ht-­mNG. (A) Brief illustration of in vitro evaluation with confocal scanning microscopy and flow cytometry.
NIH3T3 cells (unrelated control), bEnd.3 cells (BBB model), C6 cells (glial model), and HT22 cells (hippocampal neuron model) were used in this study. (B and C) Confocal
micrographs of cells treated with Ht-­mNG (25 μg/mL) for 120 min at 37 °C. Prior to observation, the cells were washed thrice with glycine–HCl buffer and once with PBS
for optimal fluorescence investigation. Panel (B) compares cellular uptake of all Ht-­mNG by bEnd.3 and HT22 cells. Panel (C) compares uptake of HtB-­and HtD-­mNG
between all four cell types. (D and E) Histograms and geometric MFI, respectively, generated from flow cytometry after treatment with Ht-­mNG at 25 μg/mL at 37 °C. (F)
Brief illustration of the SPR assay. Purified IR with α and β chains was immobilized on the CM5 sensorchip, and various concentrations of Ht-­mNG were injected into the
flow as analytes. Intermolecular binding was detected as mass change on the sensorchip. (G) Binding sensorgrams after Ht-­mNG was injected to the IR –immobilized
flow cell, where specific binding was expressed by subtracting nonspecific binding to blank flow cell from the total binding. (H) Comparison of cellular uptake of
Ht-­mNG by HT22 and N2a cells. (I) Brief illustration of ICV administration to mice. Mice brains were isolated 30 min after ICV administration of Ht-­mNG (250 μg/mL,
10 μL/30 g mouse). Frozen tissue slices were treated with primary antibodies to neuronal nucleic marker (NeuN) and stained with secondary antibodies conjugated to
fluorescent dyes. (J) Representative images (from three mice for each group) of hippocampal regions after ICV administration of Ht-­mNG. The bars indicate 500 μm.

injected. Unexpectedly, fusion with incomplete forms of insulin maximal RU (Rmax) were calculated. The lowest Kd for HtB-
(A and B chains of insulin) provided mNG the ability to bind IR. mNG (3.330 μM) among fusion proteins suggested possible spec­
Based on affinity analysis, the dissociation constant (Kd) and ificity to IR –expressing cells such as hippocampal neurons, and

6 of 12 https://doi.org/10.1073/pnas.2407936121 pnas.org
 the highest Rmax (152.4 RU) could be associated with the endocytosis. To examine the involvement of IR as a possible
enhanced cellular uptake of HtB-mNG. determinant, HT22 cells were incubated with high concentration
Because HT22 cells had the largest capacity to internalize insu­ of insulin (100 μg/mL) and then treated with Ht-­mNG, where
lin and Ht-mNG in the cells (Figs. 2F and 4E), these in vitro data receptor should be occupied with pretreated insulin. Contrary to
strongly reinforced the potential of insulin to deliver protein drugs expectation, in the presence of excess insulin, Ht-­mNG uptake
to hippocampal neurons. However, HT22 cells are immortalized by HT22 cells was partially elevated (Fig. 5C), suggesting that
and may not completely reflect the in vivo properties of hippocam­ uptake of Ht-­mNG cannot be competed with insulin and that
pal neurons. Therefore, we characterized rat hippocampal primary Ht-­mNG can be transferred into cells via receptor-­independent or
neurons and examined the accumulation of Ht proteins in these indirect receptor-­dependent pathways. We confirmed that insulin
primary cells. The expression of IR was similar between HT22 itself (FITC-­insulin) is internalized in a receptor-­independent
cells and hippocampal primary neurons (SI Appendix, Fig. S5 A manner (Fig. 5D), suggesting that insulin signaling (IRS/PI3K/
and B). In addition, IGF-1R was not detected in both cell lysates. Akt) for glucose metabolism and insulin internalization, possibly
Despite similar receptor expression, Ht-mNG (HtB- and for memory and learning, in the brain occur separately (24, 35).
HtD-mNG) were more effectively merged with hippocampal pri­ This possibility was confirmed by testing the hypoglycemic effect
mary neurons than HT22 cells (SI Appendix, Fig. S5C). Because of insulin and Ht-­mNG and phosphorylation of IR via insulin
primary neurons are nonproliferative, enough cells for flow cytom­ stimulation of HEK293T cells, which abundantly express IR
etry were not obtained. Instead, the effect of Ht-fusion on the (SI Appendix, Fig. S1 A and B). Although subcutaneous human
uptake by hippocampal primary neurons was quantitatively ana­ insulin injection (1 IU/kg body weight) in mice showed strong
lyzed with Ht-Nluc. HtB-fusion effectively enhanced Nluc uptake hypoglycemic effects, HtB-­and HtD-­mNG did not affect blood
by primary neurons, but the quantitative comparison between the glucose levels (SI Appendix, Fig. S6A). The gradual elevation of
HT22 cell and hippocampal primary neuron might not be precise blood glucose levels after injection was normal reaction of mice due
because of the low content and limited detection of nonprolifer­ to the stress during experimental operation (22). Western blotting
ating primary neurons (SI Appendix, Fig. S5D). The specificity of demonstrated that IR in HEK293T cells was phosphorylated by
Ht-mNG accumulation to hippocampal neurons was further adding human insulin consisting of both A and B chains, but no
investigated by comparing their uptake by HT22 and N2a cells. effect of only A or B chain and Ht-­fusion proteins was observed
HtB- and HtD-mNG were efficiently taken up by HT22 cells, (SI Appendix, Fig. S6 B and C).
but not by N2a cells (Fig. 4H), suggesting the potential of Ht The contribution of IR to the specific accumulation of Ht pro­
proteins to specifically target to the hippocampal neurons. teins was further investigated with IR-overexpressing HT22 cells
(Fig. 5E). We cloned the plasmid for expressing IR linked to a
Hippocampal Distribution of Ht Proteins after ICV Admin­ fluorescent tag (DsRed) in HT22 cells. The cells expressed IR-
istration. Based on the results of in vitro examinations (Fig. 4 DsRed after transfection with reagents (TransIT-X2 and TransIT
B, C, and E), we tested the capacity of Ht proteins in animals. 293) for 24 or 48 h, with low cytotoxicity (SI Appendix, Fig. S7
Downloaded from https://www.pnas.org by 190.6.179.10 on January 10, 2025 from IP address 190.6.179.10.

Mice received Ht-­mNG via ICV injection, and their distribution A and B). We separated DsRed-positive (IR-overexpressing) and
in the brain was evaluated with confocal microscopy of brain DsRed-negative (wild-type) cells by flow cytometry and compared
sections (Fig. 4I). The molecules administered via ICV injection the cellular internalization of Ht-mNG (Fig. 5F). Only HtD-mNG
could rapidly reach neighboring hippocampal regions; however, internalized more effectively into DsRed-positive cells than
normally, most molecules cannot penetrate the neuronal cell DsRed-negative cells, but no statistically significant increase was
layer. Although mNG without insulin was located outside the observed for other Ht-mNG (Fig. 5G). Because FITC-insulin was
neuronal cell layer, HtA-­and HtB-­tagged mNG could accumulate more efficiently taken up by DsRed-positive cells (Fig. 5H), the
in neuronal cells, including axonal parts (Fig. 4I). Contrary to contribution of IR could be assessed with this system. These results
expectation, full-­length insulin (HtD-­mNG) was less effective suggest that accumulation of HtA- and HtB-mNG is IR inde­
in delivering mNG to neuronal cells, possibly because its large pendent, although the SPR assay demonstrated the binding of
molecular weight reduced hippocampal diffusion. Ht-mNG—even incomplete insulin forms A- and B-chains—to
IR. Confocal microscopy could not reveal the colocalization of
Mechanisms for Accumulation of Ht Proteins in the Hippocampal Ht-mNG with IR-DsRed in cells (Fig. 5I), strengthening the
Neurons. Although the expression of IR on hippocampal neurons hypothesis that IR does not contribute to the cellular accumula­
may be associated with the specificity of cell accumulation of tion of Ht-mNG, in particular HtA- and HtB-mNG.
insulin and Ht proteins (Figs. 2K and 4G), the detailed mechanisms To find the mechanisms for Ht-fusion-mediated neuronal cell
must be understood. To examine the involvement of energy-­ accumulation of proteins, we investigated the involvement of endo­
dependent endocytosis in the accumulation and internalization cytosis types using siRNA for knocking down dynamin-2, which is
of Ht-­mNG by HT22 cells, we compared uptake between 37 °C the main protein for clathrin-, caveolae-, and receptor-mediated
and 4 °C (Fig. 5A). The cells were washed with acidic glycine–HCl endocytosis, and the macropinocytosis inhibitor, rottlerin (Fig. 5J)
buffer or neutral PBS to separate the fraction attached on the (36–41). The expression of dynamin-2 in HT22 cells was reduced
cell surface from the fraction internalized by the cells. We have by treatment with siRNA for dynamin-2 (siDNM2) (SI Appendix,
reported that glycine–HCl buffer is most effective for detaching Fig. S7 C and D). Because the transfection reagent or siRNA treat­
fractions from the cell surface and maximizing fluorescence ment was potentially cytotoxic, Ht-mNG uptake was compared
intensity inside the cells (34). Strong fluorescence of HtB-­mNG between control siRNA and siDNM2-transfected cells (SI Appendix,
and HtD-­mNG in HT22 cells was abrogated by incubation at Fig. S7E). No reduction in the internalization of Ht-mNG was
4 °C and washing with glycine–HCl buffer (Fig. 5 B, Upper). Flow observed in siDNM2-transfected HT22 cells (Fig. 5K). The treat­
cytometry confirmed the reduced fluorescence of cells treated with ment of HT22 cells with rottlerin (20 μM) drastically decreased the
Ht-­mNG at 4 °C (Fig. 5C). Even after washing with PBS, the internalization of Ht-mNG (Fig. 5L). FITC-insulin was less effi­
internalization of Ht-­mNG was almost abolished at 4 °C (Fig. 5 ciently internalized into rottlerin-treated HT22 cells, compared with
B, Lower), suggesting that Ht-­mNG is mainly internalized by untreated cells (Fig. 5M), which is consistent with our recent pub­
fluid-­phase endocytosis and not cell surface receptor-­mediated lication (37). By contrast, treatment of cells with a macropinocytosis

PNAS 2024 Vol. 121 No. 41 e2407936121 https://doi.org/10.1073/pnas.2407936121 7 of 12

 A B

C D

E F G H

I
Downloaded from https://www.pnas.org by 190.6.179.10 on January 10, 2025 from IP address 190.6.179.10.

J K L M

Fig. 5.   Mechanistic evaluation of the cellular accumulation and uptake of Ht-­mNG and FITC-­insulin. (A) Schematic illustration of the experimental conditions for
examining involvement of energy-­dependent endocytosis and specific cell surface receptor. (B) Confocal micrographs of HT22 cells after incubation with Ht-­mNG
(25 μg/mL) at 37 °C or 4 °C followed by washing with glycine–HCl buffer or PBS. (C) MFI of HT22 cells calculated from flow cytometry after incubation with Ht-­mNG
at 37 °C or 4 °C or in the presence of excessive concentration of unlabeled insulin (100 μg/mL). (D) MFI of HT22 cells after incubation with FITC-­insulin at 37 °C in the
presence or absence of unlabeled insulin. (E) Illustration of IR-­overexpressing cells for determining the contribution of IR to Ht-­fusion proteins. IR was transiently
expressed in HT22 cells transfected with the IR-­encoding plasmid. In this plasmid, DsRed is tagged at the 3′ end (C terminus) of the IR for detecting IR-­overexpressing
cells. (F) Dot plot from flow cytometry for gating IR-­DsRed-­positive and -­negative cells. (G and H) MFI from flow cytometry after incubation with Ht-­mNG and FITC-­
insulin, respectively. Fluorescence intensity was compared between DsRed-­positive and -­negative cells. (I) Confocal micrographs of IR-­DsRed-­transfected HT22 cells
and wild-­type (WT) HT22 cells after incubation with Ht-­mNG or FITC-­insulin. Two micrographs shown for untreated control cells Dulbecco’s modified Eagle medium
(DMEM) were taken from the same samples, but detection settings were differently optimized for comparison with mNG and FITC-­insulin samples. (J) Schematic
illustration of the experiment for evaluating involvement of endocytosis pathways. siRNA was used for knocking down the major protein (dynamin) associated
with clathrin-­mediated endocytosis, and rottlerin (20 μM) was used for reducing macropinocytosis. (K) Flow cytometric analysis of dynamin-­deleted HT22 cells after
incubation with Ht-­mNG. (L and M) Flow cytometric analysis of rottlerin-­treated HT22 cells after incubation with Ht-­mNG or FITC-­insulin, respectively.

8 of 12 https://doi.org/10.1073/pnas.2407936121 pnas.org
 marker (FD70) clarified that the activity of macropinocytosis in In experiments with single culture, the accumulation of these
HT22 cells was not intrinsically high compared with other cells mutants was not specific to HT22 cells compared with other cells
(3T3, bEnd.3, and C6 cells) (SI Appendix, Fig. S8 A and B). The (3T3, bEnd.3, and C6 cells) (Fig. 6E). By contrast, uptake by
uptake of FD70 by HT22 cells was reduced at 4 °C and in the HT22/C6 cells in coculture Transwell systems showed a tendency
presence of rottlerin (SI Appendix, Fig. S8 C and D). Interestingly, of specific accumulation of M3-mNG to HT22 cells (SI Appendix,
FD70 uptake was facilitated the by addition of excessive insulin in Fig. S4D). SPR analysis clarified that M3- and M6-mNG could
concentration-dependent and saturable manner (SI Appendix, bind IR, although the amino acid sequences within these peptides
Fig. S8 D and E), but not increased in IR-DsRed-positive cells with­ were mostly modified from original human insulin (Fig. 6G). The
out receptor stimulation (SI Appendix, Fig. S8F). These findings binding of M3- and M6-mNG did not contribute to IR phos­
suggest that insulin and Ht-mNG were internalized into HT22 cells phorylation (SI Appendix, Fig. S6D). Interestingly, M8 mutant-
via enhanced macropinocytosis triggered by receptor stimulation. mNG, which is unrelated to insulin, could bind IR (Fig. 6G),
Insulin is often supplemented as a growth factor for cell culture and indicating its effect on enhanced uptake of M8-mNG by HT22
can potentially activate macropinocytosis as biological response, like cells. The uptake study with IR-DsRed-overexpressing HT22 cells
other growth factors (42). HtB-mNG might unlock the maximal showed that HtB-mNG and M8-mNG were similarly taken up
potential of macropinocytosis in HT22 cells; therefore, the addi­ by cells, and no differences between DsRed-positive and -negative
tional internalization of HtB-mNG might be difficult to detect even cells suggest that their addition led to maximal macropinocytosis
when overexpressing IR (Fig. 5G). These in vitro data suggest that (Fig. 6H). By contrast, internalization of M3-mNG was much
enhanced macropinocytosis of neuronal cells through exposure to higher than HtB- and M8-mNG and not enhanced in DsRed-
insulin or Ht-fusion proteins contributes to accumulation in hip­ positive cells, indicating the involvement of additional or other
pocampal neurons in vivo. mechanisms than IR-mediated macropinocytosis. Compared with
HtB-mNG uptake, M3-mNG uptake was highly saturable with
Dominant Amino Acid Sequences in Insulin for Hippocampal increase in concentration (Fig. 6I). This suggests that other poten­
Neuronal Delivery. We determined the amino acid sequence in tial receptors or shuttle proteins were involved in the neuronal
insulin necessary for hippocampal neuronal targeting. We generated internalization of M3-mNG. The examination of internalization
various insulin mutants (M1 to M6) from native insulin A and B pathways clarified that the uptake of M3- and other mutant-mNG
chains (Fig. 6A). M7 and M8 were prepared as nonspecific cell-­ were temperature dependent and rottlerin sensitive (Fig. 6 J and
penetrating peptide (penetratin) and random sequence peptide with K). Treatment with excessive insulin did not affect the uptake of
30 amino acids (two cysteines remained), respectively. Both one-­ M3- and M6-mNG (Fig. 6K), suggesting that IR-mediated mac­
third reversed (M1) and scrambled (M2) mutants of insulin A chain ropinocytosis was fully activated with M3-mNG even without
effectively increased mNG uptake by HT22 cells (Fig. 6B). Similarly, extra stimulation and contributed less to M6-mNG.
the effect of fusion with one-­third reversed (M3) and scrambled Thus, insulin mutant M3 was most efficient in delivering pro­
(M4) mutants of insulin B chain did not decrease compared with teins to hippocampal neurons. Therefore, we tested the in vivo
Downloaded from https://www.pnas.org by 190.6.179.10 on January 10, 2025 from IP address 190.6.179.10.

native HtB-­fusion, considering interfered fluorescence intensity performance of M3-mNG for hippocampal neuronal targeting
(Fig. 6 B and C). Surprisingly, M3-­fusion was approximately five after ICV administration. Fig. 6L shows the distribution of
times effective in enhancing mNG uptake compared with native M3-mNG around the hippocampal region at 30 and 60 min after
HtB-­fusion. Comparison with HT22 and N2a cells showed the ICV administration. M3-mNG-derived fluorescence accumulated
potential of M3-­fusion to deliver proteins specifically to the in the neuronal layers, suggesting that M3 can specifically deliver
hippocampal neurons (Fig. 6D). Random peptide M8 was effici­ proteins to hippocampal neurons.
ently taken up by HT22 cells, thus the complete sequence of insulin
might not be essential as a critical determinant. Unlike one-­third Noninvasive and Direct Nose-­to-­Brain Delivery and Hippocampal
reversed and scrambled mutants, the cysteine-­substitution mutants Neuronal Targeting of Ht Proteins. The in vivo studies with Ht
(with alanine [M5] or serine [M6]) dramatically reduced HtB-­ protein (mNG) showed their potential in delivering therapeutic
mNG uptake (Fig. 6B), suggesting that the intermolecular disulfide proteins effectively and specifically to hippocampal neurons after
bond is important. ICV administration (Figs. 4 J and 6L). Non-­or less-­invasive routes
To examine the necessity of cysteine in effective internalization must be developed for targeting protein drugs to hippocampal
into hippocampal neurons, additional mutants (M9, M3v2, and neurons for the future clinical application. Nose-­to-­brain delivery
M8v2) were prepared. Cysteine-substitution mutants (M3v2 and has been well studied and is considered as an approach for the direct
M8v2) were less effective in enhancing the internalization of and effective transport of drugs, including peptides and proteins
mNG (Fig. 6E). By contrast, changing the position of cysteine (44). Nasal mucosal transport of macromolecules, such as peptides,
in insulin B chain (M9) did not affect the ability of insulin B proteins, and nucleic acids, prior to brain distribution is intrinsically
chain to enhance cellular internalization. The importance of the low. Therefore, we examined the transport efficiency of Ht-­fusion
intermolecular disulfide bond was further investigated with west­ proteins from the nasal cavity to the brain with or without epithelial
ern blotting. As shown in Fig. 6F, western blotting without permeation enhancers (SI Appendix, Fig. S9A). Nluc was used as
reducing agent clarified that HtB-, M3-, M8-, and M9-fusion cargo for considering detectability after intranasal administration.
mNG, which are highly sensitive to cellular internalization We found that the cell-­penetrating peptide penetratin could be a
(Fig. 6 B and E), made an intermolecular complex. Western useful carrier for therapeutic peptides and proteins crossing the
blotting with reduced samples revealed single bands of each nasal epithelial barrier (21, 27, 45, 46). We mixed HtB-­Nluc
fusion mNG, suggesting that the complex dissociated by reduc­ with penetratin and administered both intranasally to mice.
ing intermolecular disulfide bonds (Fig. 6F). Thus, inclusion of Coadministration with penetratin significantly enhanced the nose-­
cysteine in carrier peptides and the resultant intermolecular to-­brain transfer of HtB-­Nluc (SI Appendix, Fig. S9B). The highest
disulfide bonds can promote the delivery of Ht proteins to hip­ luminescence activity of HtB-­Nluc was detected in the olfactory bulb,
pocampal neurons. A recent study also reported that multivalent suggesting direct nose-­to-­brain transport via the space surrounding
insulin nanocomplex showed higher activation of IR compared olfactory neurons. By contrast, HtB-­Nluc was efficiently absorbed
to monovalent insulin (43). into the blood because the lamina propria has abundant microvessels.

PNAS 2024 Vol. 121 No. 41 e2407936121 https://doi.org/10.1073/pnas.2407936121 9 of 12

 A B C

D E F G

H I J
Downloaded from https://www.pnas.org by 190.6.179.10 on January 10, 2025 from IP address 190.6.179.10.

K L

Fig. 6.   Determination of important amino acid sequences in insulin and identifying more powerful mutants for delivering protein drugs to hippocampal neurons.
(A) List of insulin mutants and their amino acid sequences. M1 and M2 are insulin A mutants, and M3 to M6 are insulin B mutants. M7 and M8 are typical strong CPP
(penetratin) and random peptides with 30 amino acids, respectively. M9, M3v2, and M8v2 are additional mutants from original insulin B, M3, and M8, respectively.
(B) MFI from flow cytometry of HT22 cells incubated with insulin mutant-­mNG for 2 h at 37 °C. (C) Variation in fluorescence intensity of insulin mutant-­mNG at the
same concentration (10 μg/mL). (D) Comparison of insulin mutant-­mNG uptake by NIH3T3, bEnd.3, C6, and HT22 cells with flow cytometry. (E) Comparison of cellular
uptake of M3-­mNG by HT22 and N2a cells. (F) Western blotting of insulin mutant-­mNG. Samples were loaded at 75 ng/mL of purified protein. PVDF membranes
were treated with antibodies for mNG. Arrows indicate the molecular weights of original mNG. (G) Binding sensorgrams after insulin mutant-­mNG was injected to
IR –immobilized flow cell, where specific binding was expressed by subtracting nonspecific binding to blank flow cell from total binding. (H) Uptake by HT22 cells
overexpressing IR-­DsRed. (I) Concentration dependence of HtB-­ and M3-­mNG on HT22 cell uptake. (J) Uptake of insulin mutant-­mNG by HT22 cells at 37 °C and
4 °C. (K) Uptake by HT22 cells treated with rottlerin (20 μM) or unlabeled insulin (100 μg/mL). (L) Representative images (from three mice for each time point) around
hippocampal regions at 30 and 60 min after ICV administration of the insulin mutant (M3)-­mNG in mice. The slices prepared from frozen brain specimens were
stained with primary antibodies against NeuN and secondary antibodies with fluorescent dye (AF647). The bars indicate 500 μm.

Importantly, HtB-­Nluc could be distributed throughout the brain, hippocampus should be a therapeutic target when neurodegenerative
including the deepest part––hippocampus (SI Appendix, Fig. S9B). disorders, including AD, impair memory and learning. Delivering
Thus, protein drugs can be noninvasively targeted to hippocampal drugs that modify Aβ and tau, as well as repair neurons, to appro­
neurons with the Ht-­fusion strategy. priate parts of the brain is important for effective causal AD therapy.
In this study, insulin could be localized in hippocampal neurons after
Discussion effective nose-to-brain delivery (Fig. 2B), consistent with our recent
findings (27). The specificity of insulin accumulation in the brain
The hippocampus is a functionally and positionally central was related to the level of IR expression on the cells (Fig. 2 F and K).
part of the brain, contributing to memory and learning. The Therefore, we designed Ht proteins that target the hippocampus.

10 of 12 https://doi.org/10.1073/pnas.2407936121 pnas.org
 In vitro and in vivo data demonstrated that Ht proteins could be expression, and all other in vitro and in vivo biological analyses. Key protocols
delivered preferentially to hippocampal neurons by binding the IR used in this study and ethical issues are summarized below, with fully detailed
(Fig. 4). Importantly, even incomplete forms of insulin (only B descriptions present in SI Appendix.
chain) could bind the IR and deliver cargo. In neuronal cells, insulin
Cell Culture. Human embryonic kidney HEK293T (CTL-­3216), mouse cerebral
and Ht proteins induced macropinocytosis through IR binding
cortex endothelial bEnd.3 (CRL-­2299), and rat glial fibroblast C6 (CCL-­107) cells
(Fig. 5). Increased macropinocytosis, possibly actin-dependent neu­ were purchased from the American Type Culture Collection (Rockville, MD, USA).
ronal ultrafast endocytosis (47, 48), contributed to effective delivery Mouse hippocampal neuronal HT22 cells (SCC129) were purchased from Merck
to hippocampal neurons in vivo. Although the substitution of KGaA (Darmstadt, Germany). NIH3T3 and N2a cells were purchased from the
cysteines in Ht proteins decreased their neuronal uptake, the partial RIKEN Bioresource Research Center (Tsukuba, Japan). For detail, see SI Appendix.
reversal of the amino acid sequence within the insulin B chain unex­
pectedly boosted cellular uptake by approximately fivefold, showing Generation of Constructs. The genetic engineering experiments were
the in vivo performance of hippocampal neuronal targeting (Fig. 6). approved by the Institutional Review Board of Kobe Gakuin University (protocol
Thus, fusion or conjugation with insulin (even a single chain) or codes: 22-­01, 22-­07, and 23-­03). Original vectors containing mNG (pNCS-­mNG)
its mutants can be a revolutionary approach for delivering therapeu­ and Nluc (N1361, pNLF1) were purchased from Allele Biotechnology (San Diego,
tic proteins to hippocampal neurons. This is potentially applicable CA, USA) and Promega Corp. (Madison, WI, USA), respectively. The DNA fragments
to antibodies and neurotrophic factors, as well as low molecular of preproinsulin including B chain, c-­peptide, and A chain, were synthesized
weight medicines. In recent years, neuronal targeting with Rabies by Integrated DNA Technologies, Inc. (Coralville, IA, USA). PCR and sequencing
virus glycoprotein derived short peptide (RVG29) has been most primers were synthesized by Eurofins Genomics, Inc. (Tokyo, Japan). Fragments
encoding insulin mutants were synthesized by Integrated DNA Technologies or
studied for effective neuronal delivery of proteins and nucleic acids
Eurofins. For detail, see SI Appendix.
(49, 50). However, RVG29 is not specific for delivering to hippocam­
pal neurons (51), and virus-derived carriers are unfavorable consid­ Protein Expression. The constructs encoding insulin or mutated insulin-­
ering their clinical use. Although studies have shown efficient genome Nluc or mNG fusion proteins were transformed into BL21(DE3) competent
editing in hippocampal neurons with nonviral targeted axonal cells (#69450, Merck KGaA). HtD-­Nluc and HtD-­mNG were transformed into
import peptide carrier, which was identified by phage display screen­ SHuffle T7 Express competent cells (C3029J, New England Biolabs). For detail,
ing (52), this peptide was originally studied as a carrier to spinal cord see SI Appendix.
motor neurons (53). Another study suggested that lipid nanoparticles
Animals. All animal studies were performed at Kobe Gakuin University and
(LNP) could deliver siRNAs into cultured neurons to achieve tar­
complied with the regulations of the Committee on Ethics in the Care and Use of
geted gene silencing (54). However, the silencing effect was based
Laboratory Animals. The animal experiments were approved by the Institutional
on the specific action of siRNAs against neuronal genes, and the Review Board of Kobe Gakuin University (protocol codes: A22-­02, A22-­03, A23-­
LNP themselves were located widely in the brain. Unlike other tech­ 27, and A23-­28). Male ddY mice (6 wk old, 30 g body weight) were purchased
niques, our hippocampal neuron–targeting technology (or HiNT) from Japan SLC (Shizuoka, Japan). For detail, see SI Appendix.
can deliver drugs more safely and specifically to hippocampal neu­
Downloaded from https://www.pnas.org by 190.6.179.10 on January 10, 2025 from IP address 190.6.179.10.

rons. Importantly, HiNT will contribute to maximizing the thera­ Data, Materials, and Software Availability. All study data are included in the
peutic potential of various biological drugs for AD and other article and/or SI Appendix.
neurodegenerative disorders. Because intranasal coadministration
ACKNOWLEDGMENTS. This study was supported in part by JSPS KAKENHI
with the membrane permeation enhancer, penetratin, might be capa­
for Scientific Research B (grant numbers 23H03751 and 23K28439), Mochida
ble of delivering Ht proteins (Nluc) to the brain, less- or noninvasive
Memorial Foundation for Medical and Pharmaceutical Research, and Takeda
and convenient administration will be more optimized for clinical Science Foundation. We are grateful to Ms. Mai Itagaki and Mr. Yoshinori Nasu
application of HiNT in the future. (DDS Laboratory, Kobe Gakuin University, Japan) for their technical help in this
study.
Materials and Methods
Detailed materials and methods are provided in SI Appendix, Materials and Author affiliations: aLaboratory of Drug Delivery Systems, Faculty of Pharmaceutical
Methods, including all materials and instruments, genetic engineering, protein Sciences, Kobe Gakuin University, Chuo-­ku, Kobe, Hyogo 650-­8586, Japan

1. A. L. Boxer, R. Sperling, Accelerating Alzheimer’s therapeutic development: The past and future of 14. A. Arguello et al., Molecular architecture determines brain delivery of a transferrin receptor-­targeted
clinical trials. Cell 186, 4757–4772 (2023). lysosomal enzyme. J. Exp. Med. 219, e20211057 (2022).
2. C. H. van Dyck et al., Lecanemab in early Alzheimer’s disease. N. Engl. J. Med. 388, 9–21 (2023). 15. T. Okuyama et al., A phase 2/3 trial of Pabinafusp Alfa, IDS fused with anti-­human transferrin
3. L. S. Honig et al., ARIA in patients treated with lecanemab (BAN2401) in a phase 2 study in early receptor antibody, targeting neurodegeneration in MPS-­II. Mol. Ther. 29, 671–679 (2021).
Alzheimer’s disease. Alzheimers Dement. N9, e12377 (2023). 16. P. Zhao et al., A tetravalent TREM2 agonistic antibody reduced amyloid pathology in a mouse model
4. C. J. Mummery et al., Tau-­targeting antisense oligonucleotide MAPT(Rx) in mild Alzheimer’s of Alzheimer’s disease. Sci. Transl. Med. 14, eabq0095 (2022).
disease: A phase 1b, randomized, placebo-­controlled trial. Nat. Med. 29, 1437–1447 (2023). 17. B. van Lengerich et al., A TREM2-­activating antibody with a blood-­brain barrier transport vehicle
5. J. W. Vogel et al., Spread of pathological tau proteins through communicating neurons in human enhances microglial metabolism in Alzheimer’s disease models. Nat. Neurosci. 26, 416–429 (2023).
Alzheimer’s disease. Nat. Commun. 11, 2612 (2020). 18. J. Niewoehner et al., Increased brain penetration and potency of a therapeutic antibody using a
6. Z. Wu et al., Neurotrophic signaling deficiency exacerbates environmental risks for Alzheimer’s monovalent molecular shuttle. Neuron 81, 49–60 (2014).
disease pathogenesis. Proc. Natl. Acad. Sci. U.S.A. 118, e2100986118 (2021). 19. D. Correa et al., Intranasal delivery of full-­length anti-­Nogo-­A antibody: A potential alternative
7. Y. Fukushima et al., Treatment of ischemic neuronal death by introducing brain-­derived route for therapeutic antibodies to central nervous system targets. Proc. Natl. Acad. Sci. U.S.A. 120,
neurotrophic factor mRNA using polyplex nanomicelle. Biomaterials 270, 120681 (2021). e2200057120 (2023).
8. A. H. Nagahara et al., Early BDNF treatment ameliorates cell loss in the entorhinal cortex of APP 20. A. E. Aly et al., Delivery of mutant huntingtin-­lowering antisense oligonucleotides to the brain by
transgenic mice. J. Neurosci. 33, 15596–15602 (2013). intranasally administered apolipoprotein A-­I nanodisks. J. Control Release 360, 913–927 (2023).
9. A. H. Nagahara et al., Neuroprotective effects of brain-­derived neurotrophic factor in rodent and 21. N. Kamei, M. Takeda-­Morishita, Brain delivery of insulin boosted by intranasal coadministration
primate models of Alzheimer’s disease. Nat. Med. 15, 331–337 (2009). with cell-­penetrating peptides. J. Control Release 197, 105–110 (2015).
10. G. C. Terstappen, A. H. Meyer, R. D. Bell, W. Zhang, Strategies for delivering therapeutics across the 22. N. Kamei et al., Effective nose-­to-­brain delivery of exendin-­4 via coadministration with cell-­
blood-­brain barrier. Nat. Rev. Drug Discov. 20, 362–383 (2021). penetrating peptides for improving progressive cognitive dysfunction. Sci. Rep. 8, 17641 (2018).
11. W. M. Pardridge, A historical review of brain drug delivery. Pharmaceutics 14, 1283 (2022). 23. K. Talbot et al., Demonstrated brain insulin resistance in Alzheimer’s disease patients is associated
12. Y. J. Zuchero et al., Discovery of novel blood-­brain barrier targets to enhance brain uptake of with IGF-­1 resistance, IRS-­1 dysregulation, and cognitive decline. J. Clin. Invest. 122, 1316–1338
therapeutic antibodies. Neuron 89, 70–82 (2016). (2012).
13. T. Okuyama et al., Iduronate-­2-­sulfatase with anti-­human transferrin receptor antibody for 24. S. M. Gray, K. W. Aylor, E. J. Barrett, Unravelling the regulation of insulin transport across the brain
neuropathic mucopolysaccharidosis II: A phase 1/2 trial. Mol. Ther. 27, 456–464 (2019). endothelial cell. Diabetologia 60, 1512–1521 (2017).

PNAS 2024 Vol. 121 No. 41 e2407936121 https://doi.org/10.1073/pnas.2407936121 11 of 12

 25. S. Craft et al., Safety, efficacy, and feasibility of intranasal insulin for the treatment of mild cognitive 40. E. Cocucci, R. Gaudin, T. Kirchhausen, Dynamin recruitment and membrane scission at the neck of a
impairment and alzheimer disease dementia: A randomized clinical trial. JAMA Neurol. 77, clathrin-­coated pit. Mol. Biol. Cell 25, 3595–3609 (2014).
1099–1109 (2020). 41. M. Kaksonen, A. Roux, Mechanisms of clathrin-­mediated endocytosis. Nat. Rev. Mol. Cell Biol. 19,
26. N. Kamei et al., Effect of an enhanced nose-­to-­brain delivery of insulin on mild and progressive 313–326 (2018).
memory loss in the senescence-­accelerated mouse. Mol. Pharm. 14, 916–927 (2017). 42. I. Nakase, N. B. Kobayashi, T. Takatani-­Nakase, T. Yoshida, Active macropinocytosis induction by
27. N. Kamei et al., Investigation of the transport pathways associated with enhanced brain delivery of stimulation of epidermal growth factor receptor and oncogenic ras expression potentiates cellular
peptide drugs by intranasal coadministration with penetratin. Pharmaceutics 13, 1745 (2021). uptake efficacy of exosomes. Sci. Rep. 5, 10300 (2015).
28. D. Wu, J. Yang, W. M. Pardridge, Drug targeting of a peptide radiopharmaceutical through the 43. J. Spratt et al., Multivalent insulin receptor activation using insulin-­DNA origami nanostructures.
primate blood-­brain barrier in vivo with a monoclonal antibody to the human insulin receptor. Nat. Nanotechnol. 19, 237–245 (2024).
J. Clin. Invest. 100, 1804–1812 (1997). 44. M. L. Borrajo, M. J. Alonso, Using nanotechnology to deliver biomolecules from nose to brain–
29. A. Kleinridders, H. A. Ferris, W. Cai, C. R. Kahn, Insulin action in brain regulates systemic metabolism Peptides, proteins, monoclonal antibodies and RNA. Drug Deliv. Transl. Res. 12, 862–880 (2022).
and brain function. Diabetes 63, 2232–2243 (2014). 45. E. S. Khafagy et al., Systemic and brain delivery of leptin via intranasal coadministration with
30. M. Soto, W. Cai, M. Konishi, C. R. Kahn, Insulin signaling in the hippocampus and amygdala cell-­penetrating peptides and its therapeutic potential for obesity. J. Control Release 319, 397–406
regulates metabolism and neurobehavior. Proc. Natl. Acad. Sci. U.S.A. 116, 6379–6384 (2019). (2020).
31. M. P. Hall et al., Engineered luciferase reporter from a deep sea shrimp utilizing a novel 46. N. Kamei et al., Therapeutic effects of anti-­amyloid beta antibody after intravenous injection and
imidazopyrazinone substrate. ACS Chem. Biol. 7, 1848–1857 (2012). efficient nose-­to-­brain delivery in Alzheimer’s disease mouse model. Drug Deliv. Transl. Res. 12,
32. N. C. Shaner et al., A bright monomeric green fluorescent protein derived from Branchiostoma 2667–2677 (2022).
lanceolatum. Nat. Methods 10, 407–409 (2013). 47. S. Watanabe et al., Ultrafast endocytosis at mouse hippocampal synapses. Nature 504, 242–247
33. J. Lobstein et al., SHuffle, a novel Escherichia coli protein expression strain capable of correctly (2013).
folding disulfide bonded proteins in its cytoplasm. Microb. Cell Fact. 11, 56 (2012). 48. T. H. Ogunmowo et al., Membrane compression by synaptic vesicle exocytosis triggers ultrafast
34. N. Kamei et al., Optimization of the method for analyzing endocytosis of fluorescently tagged endocytosis. Nat. Commun. 14, 2888 (2023).
molecules: Impact of incubation in the cell culture medium and cell surface wash with glycine-­ 49. P. Kumar et al., Transvascular delivery of small interfering RNA to the central nervous system.
hydrochloric acid buffer. J. Control Release 310, 127–140 (2019). Nature 448, 39–43 (2007).
35. M. Konishi et al., Endothelial insulin receptors differentially control insulin signaling kinetics in 50. L. Alvarez-­Erviti et al., Delivery of siRNA to the mouse brain by systemic injection of targeted
peripheral tissues and brain of mice. Proc. Natl. Acad. Sci. U.S.A. 114, E8478–E8487 (2017). exosomes. Nat. Biotechnol. 29, 341–345 (2011).
36. J. J. Rennick, A. P. R. Johnston, R. G. Parton, Key principles and methods for studying the endocytosis 51. S. S. Kim et al., Targeted delivery of siRNA to macrophages for anti-­inflammatory treatment.
of biological and nanoparticle therapeutics. Nat. Nanotechnol. 16, 266–276 (2021). Mol. Ther. 18, 993–1001 (2010).
37. M. Itagaki, Y. Nasu, C. Sugiyama, I. Nakase, N. Kamei, A universal method to analyze cellular 52. D. L. Sellers, K. Lee, N. Murthy, S. H. Pun, TAxI-­peptide targeted Cas12a ribonuclease protein
internalization mechanisms via endocytosis without non-­specific cross-­effects. FASEB J. 37, e22764 nanoformulations increase genome editing in hippocampal neurons. J. Control Release 354,
(2023). 188–195 (2023).
38. K. Sarkar, M. J. Kruhlak, S. L. Erlandsen, S. Shaw, Selective inhibition by rottlerin of macropinocytosis 53. D. L. Sellers et al., Targeted axonal import (TAxI) peptide delivers functional proteins into
in monocyte-­derived dendritic cells. Immunology 116, 513–524 (2005). spinal cord motor neurons after peripheral administration. Proc. Natl. Acad. Sci. U.S.A. 113,
39. P. Oh, D. P. McIntosh, J. E. Schnitzer, Dynamin at the neck of caveolae mediates their budding to 2514–2519 (2016).
form transport vesicles by GTP-­driven fission from the plasma membrane of endothelium. J. Cell 54. R. L. Rungta et al., Lipid nanoparticle delivery of sirna to silence neuronal gene expression in the
Biol. 141, 101–114 (1998). brain. Mol. Ther. Nucleic Acids 2, e136 (2013).
Downloaded from https://www.pnas.org by 190.6.179.10 on January 10, 2025 from IP address 190.6.179.10.

12 of 12 https://doi.org/10.1073/pnas.2407936121 pnas.org